Re: [ide-dev] -7.5% absolute, -14% relative user loss this year

[Lars Vogel <lars.vogel@xxxxxxxxxxx>]

IntelliJ has some very nice usability features. Most IntelliJ users I know prefer their IDE because of it.

Examples for better usability:

- Dark theme looks good -> we are working on improving it in Eclipse

- Highlighting braces uses a color which actually let you see the matching color -> JDT team plans to improve this in M7

- All trees are searchable by default if you simple start typing -> awesome feature I would love to have this in Eclipse but have no idea how to implement this

- Shift + Shift for open resources is so much easier to us compared to our shortcut

- Alt+Enter is also easier to reach compared with Ctrl+1

- Debugger is much better then ours it adds the value of the variable into the editor after the code. Also our default Eclipse debug perspective is IMHO not very usable.

But IntelliJ has also certain issues:

- It is relatively slow

- Impossible to open (efficiently) several projects at the same time

- Swing UI start to look broken on Linux

- Lots more but let's not tell them ;-)

Best regards, Lars

On Apr 13, 2017 4:14 PM, "Mickael Istria" <mistria@xxxxxxxxxx> wrote:

On 04/13/2017 03:56 PM, Patrik Suzzi wrote:

1. Probably we can't compete with Intelli-J on the android development.

Indeed, unless several people are ready to commit to work full time on it for several monthes.

2. We can improve Eclipse WTP to improve the web development experience. This is where we lost part of the users, in favor of webstorm and netbeans.

I fully agree. The state of JS edition got improved, but the HTML and CSS editors are still far behind many other ones in term of usability and features. However, this would require more resources on WTP, more people to review the various incoming patches ( https://git.eclipse.org/r/#/q/status:open+project:sourceediting/webtools.sourceediting ) or if the current committers cannot commit on reviewing the patches fast enough, we could recruit interested parties (such as you and Gauthier as committers).

3. We could gain users proposing a new offer: a lightweight Eclipse IDE integrated with LSP-editors and cloud IDEs. 

   - Eclipse Two is a good point to start, where we could plug CEF, (or Electron #?) and LSP4E based editors.

I'm currently only considering Eclipse IDE, not the other ones.
LSP4E is only targetting Eclipse IDE.

   - Also, Chromium Embedded will give us a chance to re-think E4 RCP applications. (A chromium-based framework, which can run RCP Plugins?)

I don't think users have left Eclipse IDE because it's not based on CEF nor Electron nor Chromium Embedded. They left because of missing or clunky high-level features that are better supported by competitors. It's not at all a matter of technology nor design IMHO, but "only" a matter of implementing and placing the right features at the right place and the right time in the user-stories.
People typically complain about completion being suboptimal, or the need to use keystrokes or certain characters to trigger completion, the lack of some refactorings, the inconsistency between Maven and JDT model... none of this is an issue that is related to the core architecture.

--
Mickael Istria
Eclipse developer for Red Hat Developers
My blog - My Tweets

_______________________________________________
ide-dev mailing list
ide-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/ide-dev